Software Engineer, Inference / Compute Infrastructure Engineering

Together AI· San Francisco, USA

Posted November 9, 2026

Apply
Experience
Not specified
Pay
$160K – $280K per year
Education
Not specified
Location
San Francisco, USA
Work mode
Hybrid
Job type
Full-time

About this role

Together AI is an AI cloud that runs inference, fine-tuning and training for AI developers. This engineer builds the Kubernetes-based control plane that sets up and runs its GPU fleet, with self-service APIs and automatic repair. Open at junior, senior and staff levels.

What you will do

  • Build the state machine for the full life of a physical host
  • Build self-service APIs so teams can create inference clusters with one call
  • Automate finding, draining and repairing bad nodes
  • Keep the provisioning pipeline reliable: retries, rollback, drift checks
  • Work with the inference and ML platform team on cluster needs
  • Treat infrastructure as software with tests, review and CI/CD

What the employer expects

  • Strong software skills in Go, Python, Rust or similar
  • Experience with durable workflow tools such as Temporal or Cadence
  • Experience with control planes or Kubernetes controllers
  • Experience with event-driven systems such as Kafka, NATS or SQS
  • Product mindset for internal platforms and APIs
  • Willing to work 4 days a week in the San Francisco office

Benefits

  • Startup equity
  • Health insurance

Skills

  • Go
  • Python
  • Rust
  • Kubernetes
  • Temporal
  • Kafka
  • GPU infrastructure
  • Event-driven systems

Apply on the employer's site

The application opens on Together AI careers page (Greenhouse). Dev Portal Notes does not collect your application or charge a fee.

Apply now

Know someone who fits this role?

Applications are handled on the employer's own site.

Prepare for this role

Close the skill gap

Some links on this page may generate a small commission at no extra cost to you.

Similar jobs

Apply for this job